Morbidity statistics by search queries
Search engines will predict the spread of cancer
Natalia Zhuravleva, Vademecum
The analysis of queries entered by users into search engines will allow predicting the spread of non-communicable diseases, such as cancer, stroke and cardiovascular diseases.
These are the results of a study published by Sveta Venkatesh and her colleagues from the Center for Data Analysis and Behavioral Models of Deakin University (Australia) in the scientific journal Journal of Epidemiology & Community Health (Nguyen et al., Web search activity data accurately predict population chronic disease risk in the USA).
"The diagnosis or suspicion of cardiovascular disease is associated with the search for symptoms of the disease, side effects of drugs and other similar data," Venkatesh was quoted by Reuters.
Venkatesh and her colleagues used an array of search queries collected by the Internet giant Google in 2014 and compared it with data from the US Centers for Disease Control and Prevention (CDC), which show the main risk factors for the development of non-communicable diseases by state. These factors include the prevalence of physical activity and smoking, as well as the number of patients with hypertension and diabetes mellitus in a particular region.
The information obtained by the researchers during the analysis of search query data has previously correlated well with CDC estimates. For example, an analysis of search queries for 2011 showed that 11.2% of Alabama residents, 9.4% of New Jersey residents and 8.1% of Nevada residents suffer from diabetes. The CDC's assessment for the same period showed 11.8% of such patients in Alabama, 8.8% in New Jersey and 10.3% in Nevada.
According to the authors of the work, their methodology will help predict the spread of non-communicable diseases in almost real time. The traditional model of statistical data collection and processing takes up to three years.
At the same time, the researchers explain that their development is not intended to replace the usual collection of statistical data, but only to supplement it at a much lower cost of research.
Portal "Eternal youth" http://vechnayamolodost.ru07.04.2015